REPRODUCTIONRESEARCH

Granulosal and thecal expression of bone morphogenetic protein- and activin-binding protein mRNA transcripts during bovine follicle development and factors modulating their expression in vitro

Claire Glister, Leanne Satchell and Phil G Knight School of Biological Sciences, University of Reading, Hopkins Building, Whiteknights, Reading RG6 6UB, UK Correspondence should be addressed to P G Knight; Email: [email protected]

Abstract

Evidence supports local roles for transforming growth factor b superfamily members including activins and bone morphogenetic proteins (BMP) in follicle development. Access of these ligands to signalling receptors is likely modulated by extracellular binding proteins (BP). In this study, we compared ex vivo expression of four BPs (chordin, gremlin, noggin and follistatin) in granulosal (GC) and theca interna (TC) compartments of developing bovine antral follicles (1–18 mm). Effects of FSH and IGF on BMP and BP expression by cultured GC, and effects of LH and BMPs on BPexpression by cultured TC were also examined. Follicular expression of all four BP transcripts was higher in GC than TC compartments (P!0.001) a finding confirmed by immunohistochemistry. Follicle category affected (P!0.01) gremlin and follistatin mRNA abundance, with a significant cell-type!follicle category interaction for chordin, follistatin and noggin. Noggin transcript abundance was lower (P!0.05) in GC of large ‘E-active’ than ‘E-inactive’ follicles while follistatin mRNA level was higher (P!0.01). FSH enhanced CYP19, FSHR, INHBA and follistatin by GC without affecting BMPor BMP–BPexpression. IGF increased CYP19 and follistatin, reduced BMP4, noggin and gremlin but did not affect chordin or FSHR mRNA levels. LH increased TC androgen secretion but had no effect on BMP or BP expression. BMPs uniformly suppressed TC androgen production whilst increasing chordin, noggin and gremlin mRNA levels up to 20-fold (P!0.01). These findings support the hypothesis that extracellular BP, mostly from GC, contribute to the regulation of intrafollicular BMP/activin signalling. Enhancement of thecal BP expression by BMP implies an autoregulatory feedback role to prevent excessive signalling. Reproduction (2011) 142 581–591

Introduction and related this to the pattern of expression of type 1 and type 2 signalling receptors. Significant cell-type and During recent years a wealth of investigation has been follicle stage-dependent changes in expression of b directed at transforming growth factor (TGFB) super- individual ligands and receptors were revealed but, in family members, in particular the largest subset within general, changes in receptor transcript abundance were the family, the bone morphogenetic proteins (BMP). relatively modest, supporting the notion that additional More than 20 BMPs have been identified in vertebrates levels of control may be operational to modulate BMP and non-vertebrate species and these multifunctional signalling at the local level. proteins contribute to the regulation of cellular prolifer- Despite the likely functional importance of BMP ation, differentiation and apoptosis in a variety of tissues, signalling within the ovary, few studies have addressed including the ovary (Miyazono et al. 2001, Chen et al. the potential role of extracellular BMP-binding proteins 2004, Shimasaki et al. 2004, Knight & Glister 2006). (BP) in modulating BMP access to their signalling A number of BMPs are expressed in theca cells (TC), receptors on different ovarian cell-types. Much of the granulosa cells (GC) and oocytes in a cell- and follicle research on BMP-BPs has been in the developmental stage-specific manner and evidence supports their biology field where their crucial roles have been involvement in the regulation of folliculogenesis, identified in establishing BMP morphogen gradients ovulation and corpus luteum function (reviews: Shimasaki responsible for embryonic patterning (Thomsen 1997, et al. 2004, Juengel & McNatty 2005, Knight & Glister Balemans & Van Hul 2002, Chen et al. 2004, Kishigami 2006). Recently, Glister et al. (2010) reported a survey of & Mishina 2005). To our knowledge no comprehensive BMP2, BMP4, BMP6 and BMP7 expression in TC and surveys of BMP-BP expression throughout ovarian GC throughout bovine antral follicular development follicle growth have yet been carried out in any species.

q 2011 Society for Reproduction and Fertility DOI: 10.1530/REP-11-0150 ISSN 1470–1626 (paper) 1741–7899 (online) Online version via www.reproduction-online.org Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access 582 C Glister and others

Similarly, information is lacking on the factors (systemic and GC is modulated by gonadotrophins or insulin-like or intraovarian) that potentially modulate expression of growth factor 1 (IGF1) and iii) examine the effect of BMP-BPs by follicle cells. As such, with the exception of exogenous BMP2, BMP4, BMP6 and BMP7 on follistatin (predominantly an activin-BP), the significance expression of BP transcripts by cultured TC. of secreted BMP-BPs as intrafollicular modulators of BMP action has received little attention. Follistatin was first isolated by Robertson et al. (1987) Results from bovine follicular fluid based on its ability to More detailed characteristics of the follicle sample set suppress FSH secretion from pituitary cells and was analysed in this study, including follicular fluid steroid subsequently shown to function by binding activin with concentrations and relative expression of GC and TC high affinity, thus neutralising its bioactivity (Nakamura ‘marker’ transcripts including FSHR, CYP19, LHCGR et al. 1990). Follistatin has also been shown to bind and CYP17, are documented in our recent companion directly to several other TGFB superfamily members, paper focusing on follicular expression of BMP ligands including inhibin (Shimonaka et al. 1991) and BMP2, and receptors (Glister et al. 2010). BMP4, BMP6, BMP7 and BMP15 (Otsuka et al. 2001, Amthor et al. 2002, Glister et al. 2004) albeit with lower affinity. Noggin, a homodimeric glycoprotein, mRNA expression profiles for gremlin, chordin, was first isolated from Xenopus embryonic tissues where follistatin and noggin in developing follicles it blocks BMP action in the dorsal organising centre mRNA transcripts for each of the four BPs were readily (Lamb et al. 1993, Smith et al. 1993). Noggin displays a detectable in all GC and TC samples with average qPCR Ct high-affinity association with BMP2 and BMP4, but less values of 26, 28, 26 and 21 for noggin, chordin, gremlin so with BMP7 and BMP6; binding directly prevents and follistatin respectively. In Fig. 1, the relative interaction of ligand with type 1 receptor (Zimmerman abundance of mRNA transcripts for all four BPs examined et al. 1996). Chordin, a cysteine-rich secreted protein, was greater in GC than TC (two-way ANOVA: effect of was also first identified in Xenopus; located in the cell-type: P!0.0001). Gremlin expression was highest in dorsal organising centre (Sasai et al. 1994, 1995). It GC of 1–2 mm follicles and declined progressively as also binds directly to BMP2, BMP4 and BMP7 to prevent follicles developed to 11–18 mm (approximately fivefold BMP–receptor interaction. However, the affinity of reduction; P!0.002); no difference in GC gremlin chordin for BMP4 is some tenfold less than its counterpart expression was evident between large oestrogen-active noggin (Piccolo et al. 1996). Gremlin belongs to a (LEA) and large oestrogen-inactive (LEI) follicles. subfamily of BMP antagonists all of whom contain a Although there was no overall effect of follicle characteristic cysteine-rich domain (CAN domain) conta- category on the relative abundance of chordin mRNA, ining a particular consensus sequence (Avsian-Kretchmer & a highly significant follicle category!cell-type Hsueh 2004). As with chordin and noggin, gremlin was first interaction (P!0.0001) justified a series of post hoc identified in a Xenopus screen for genes with dorsalising pairwise comparisons within cell-type. In GC, chordin activity (Hsu et al. 1998) and has since been shown to mRNA abundance fell (approximately sixfold; P!0.05) regulate murine limb, kidney and lung development as follicles grew from 1–2 to 7–8 mm but then increased (Zuniga et al. 1999, Khokha et al. 2003, Michos et al. again (approximately sixfold; P!0.01) between the 2004). Gremlin can bind to and inhibit BMP2 and BMP4, 9–10 and 11–18 mm (LEA) follicle categories; GC chordin but does not interact with activin (Hsu et al. 1998). expression did not differ significantly between LEA and LEI Interestingly, expression of noggin and gremlin is follicles although the mean value was 50% lower in LEI enhanced by the very BMPs they act to inhibit, indicative category. In contrast to GC, chordin expression in the TC of negative feedback loops in operation (Kameda et al. layer increased from 1–2 to 9–10 mm follicles (approxi- 1999, Pereira et al. 2000). In addition, certain isoforms of mately twofold; P!0.05), before falling (approximately follistatin, noggin and chordin bind to cell-surface threefold; P!0.01) to its lowest level in LEA follicles; TC associated heparin-sulphate proteoglycans. This associ- chordin expression was approximately fivefold higher ation may serve to limit diffusion and ‘action-range’ of (P!0.01) in LEI than in LEA follicles. either antagonist and/or ligand. It may also promote Follistatin expression patterns in GC and TC were endocytosis and lysosomal degradation of the target similar to those for chordin: in GC a progressive decrease ligand (Sugino et al. 1993, 1997, Jasuja et al. 2004). between 1–2 and 9–10 mm was followed by a large Given the current lack of information on potential increase (w14-fold; P!0.001) in LEA follicles. LEI roles of extracellular BMP-BPs in the ovary, the aims of follicles had approximately sevenfold less follistatin this study were to: i) use RT-qPCR to quantify chordin, mRNA than LEA follicles (P!0.01). In contrast to GC, noggin, gremlin and follistatin mRNA expression in follistatin expression in TC increased between 1–2 and thecal and granulosal compartments of developing 9–10 mm follicles, reaching a relative abundance bovine antral follicles; ii) investigate whether in vitro comparable with that in GC of 9–10 mm follicles. expression of the above BP transcripts by cultured TC Transcript level then fell (P!0.05) to its lowest level in

Reproduction (2011) 142 581–591 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access BMP-binding proteins and follicle development 583

Gremlin gremlin expression. Chordin and follistatin expression 30 Z ! 25 in GC were highly correlated (r 0.73; P 0.0001) and Category P<0.002 20 both were positively correlated with E2:P4 ratio and with Cell–type P<0.0001 15 Category × cell-type P=0.12 GC expression of FSHR, LHCGR and CYP19.Conversely, 10 5 GC expression of noggin was negatively correlated with 0 CYP19, LHCGR and follistatin expression. The corresponding correlation matrix for TC (Fig. 2B) 8 Chordin 7 Category P=0.26 showed fewer significant relationships than observed 6 Cell-type P<0.0001 for GC. However, there were strong positive correlations 5 Category × cell-type P<0.0001 4 between CYP17 and LHCGR (rZ0.65; P!0.0001), 3 2 between follistatin and chordin (rZ0.59; P!0.0001) 1 Granulosa Z 0 and between noggin and gremlin expression (r 0.72; Theca P!0.0001). Conversely, a weak negative correlation 14 Follistatin 12 was evident between LHCGR and noggin expression in 10 TC (rZK0.35; P!0.05). 8 Category P<0.01 6 Cell-type P<0.0001

Relative mRNA abundance Relative 4 Category × cell-type P<0.0001 2 Immunohistochemical localisation of BPs in ovary 0 sections Noggin 14 12 To reinforce the above mRNA expression data ovarian 10 Category P=0.07 8 Cell-type P<0.0001 immunohistochemical staining was carried out using anti- 6 Category × cell-type P<0.0001 bodies against each of the four BPs. As shown in Fig. 3, 4 2 immunoreactivity corresponding to all four BPs was evident 0 in follicles. In each case the intensity of staining was much 1–2 3–4 5–6 7–8 9–10 11–18 11–18 E:P>1 E:P<1 greater in the GC compartment although weak immuno- Follicle size category (mm) reactivity was also present in TC and stromal compartments. Figure 1 Changes in relative abundance of mRNA transcripts for gremlin, chordin, follistatin and noggin in thecal and granulosal Effects of FSH and IGF on expression of CYP19, FSHR, compartments of developing bovine antral follicles. Follicles in the 11–18 mm size class have been subdivided on the basis of oestrogen to INHBA, BMP and BMP-BP transcripts by cultured GC progesterone ratio (E:P ratio) as large ‘oestrogen-active’ (LEA) follicles Treatment of cultured bovine GC with an optimal dose (E:P ratio O1; open box) or large ‘oestrogen-inactive’ (LEI) follicles levelofFSH(0.33ng/ml)increasedoestradiol(E2) (E:P ratio !1; shaded box). Values are means and bars indicate S.E.M. Z secretion w40-fold (22.4G6.7 vs 0.50G0.12 ng/ml in (n 4–9). Results of two-way ANOVA are summarised. Note that the ! same sample set was used in our previous study (Glister et al. 2010)to controls; P 0.001) and promoted a marked upregulation analyse expression of BMPs, BMP receptors and various follicular of CYP19 expression (w35-fold; P!0.01) that was ‘markers’ including LHCGR, FSHR, CYP19 and CYP17. accompanied by increases in FSHR (approximately fivefold; P!0.05), follistatin (approximately sixfold; TC of LEA follicles but was approximately fourfold P!0.01) and INHBA (w11-fold; P!0.01) mRNA abun- higher in LEI than in LEA follicles (P!0.05). dance (see Fig. 4). However, FSH did not affect abundance Abundance of noggin mRNA was similarly high in GC of all follicle categories with the exception of LEA of BMP2, BMP4, BMP6, BMP7, chordin, gremlin or noggin follicles which showed a approximately tenfold transcripts. Treatment of GC with IGF increased E2 ! secretion approximately sixfold (3.27G1.08 vs 0.50 reduction (P 0.01) relative to 9–10 mm follicles and G ! LEI follicles. Levels of noggin mRNA in TC were much 0.12 ng/ml in controls; P 0.01)and,asshownin less than in the corresponding GC samples in all follicle Fig. 4, increased the abundance of transcripts for CYP19 (w12-fold; P!0.05) and follistatin (approximately three- categories except LEA follicles when transcript levels ! were similar in both cell-types. fold; P 0.05) while decreasing expression of BMP4 (approximately tenfold; P!0.01), noggin (approximately threefold; P!0.05) and gremlin (approximately threefold; Correlations amongst expression levels of different P!0.05). IGF did not affect the abundance of FSHR, transcripts in developing follicles INHBA, BMP2, BMP6, BMP7 or chordin mRNA. Figure 2A shows a correlation matrix listing the pairwise relationships between follicle diameter, E :P ratio 2 4 Effect of LH on expression of BMP and BMP-BP and relative transcript abundance of seven different transcripts by cultured TC transcripts in GC. Follicle diameter was positively corre- lated with E2:P4 ratio and with GC expression of CYP19, LH treatment resulted in a fourfold increase in A4 sec- LHCGR and follistatin, but was negatively correlated with retion by cultured TC (124.1G26.8 vs 29.5G7.0 ng/ml in www.reproduction-online.org Reproduction (2011) 142 581–591

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access 584 C Glister and others

A Granulosa layer

Diameter 1.00

E:P ratio 0.50*** 1.00

FSHR 0.16 0.36** 1.00

LHCGR 0.56*** 0.50*** 0.37** 1.00

CYP19 0.64*** 0.52*** 0.56*** 0.83*** 1.00

Gremlin (–0.49***) –0.14 0.31* –0.13 –0.19 1.00

Chordin 0.28* 0.39** 0.57*** 0.56*** 0.73*** 0.11 1.00

Follistatin 0.42** 0.47*** 0.70*** 0.55*** 0.81*** –0.05 0.74*** 1.00

Noggin –0.19 –0.23 0.18 (–0.30*) (–0.38**) 0.11 –0.17 (–0.28*) 1.00 Diameter E:P ratio FSHR LHCGR CYP19 Gremlin Chordin Follistatin Noggin Figure 2 Correlation matrices for (A) granulosal layer and (B) theca interna layer of developing follicles (nZ47*) B Theca interna layer showing associations between follicle diameter, follicular Diameter 1.00 fluid E2:P4 ratio and relative abundance of selected mRNA E:P ratio 0.41** 1.00 transcripts including the four BMP/activin-binding LHCGR (–0.32*) –0.27 1.00 proteins. Statistically significant r values are in bold with CYP17 –0.23 –0.19 0.65*** 1.00 negative correlations in parenthesis. *P!0.05, **P!0.01 Gremlin 0.17 –0.05 –0.15 –0.17 1.00 and ***P!0.001. Note that the same sample set was used Chordin 0.25 –0.23 –0.04 –0.12 0.01 1.00 in our previous study (Glister et al. 2010) to analyse Follistatin 0.13 –0.13 0.10 –0.03 –0.05 0.59*** 1.00 expression of other transcripts including the four follicular Noggin 0.30* 0.27 (–0.35*) –0.14 0.72*** –0.01 –0.15 1.00 ‘markers’ included in this correlation matrix (LHCGR, Diameter E:P ratio LHCGR CYP17 Gremlin Chordin Follistatin Noggin FSHR, CYP19 and CYP17). controls; P!0.01) but did not affect the abundance information has not yet been documented for any of transcripts for BMP2, BMP4, BMP6, BMP7, chordin, species. Interestingly, the relative abundance of all four follistatin, noggin or gremlin (data not shown). BP transcripts was greater in GC compared with TC layers throughout antral follicle development. Immuno- histochemistry confirmed that the corresponding Effects of BMPs on expression of steroidogenesis-related proteins were also more abundant in GC than TC transcripts by cultured TC compartments. In contrast, BMP4, BMP6 and BMP7 Figure 5 shows that treatment of bovine TC with BMP2, transcripts were shown to be more abundant in TC than BMP4, BMP6 and BMP7 all promoted a profound GC (Glister et al. 2010). We have previously examined suppression of CYP17 expression under both basal and the effect of exogenous BMP4, BMP6 and BMP7 on LH-stimulated conditions (w200-fold reduction; bovine GC and TC cultured in vitro (Glister et al. 2004, P!0.0001). Likewise, BMPs reduced expression of 2005) and found TC to be much more sensitive to BMP w O other key transcripts involved in the steroidogenic action (IC50 1 ng/ml), exhibiting a 100-fold decrease pathway (LHCGR, STAR, CYP11A1 and HSD3B) although in LH-induced androgen secretion at the greatest dose to much lesser extents than the suppression of CYP17. tested (50 ng/ml), compared with a more modest threefold maximal BMP-induced increase in IGF- stimulated oestrogen secretion by GC treated with Effects of BMPs on expression of BMP-BP transcripts w 50 ng/ml BMP (EC50 5 ng/ml). From the present by cultured TC findings, this relative lack of sensitivity of GC versus Figure 6 shows that BMP2, BMP4, BMP6 and BMP7 had TC to exogenous BMPs could be explained, at least in divergent effects on expression of chordin, noggin and part, by the differential expression patterns of BMP-BPs gremlin but did not modify follistatin mRNA abundance. found between the two cell-types – the greater levels of Under both basal and LH-stimulated condition levels of BPs expressed by GC antagonising the signalling activity chordin and gremlin mRNA were increased (2- to 11-fold; of exogenous BMPs to a greater extent. Thus, in an in P!0.05) by BMP2, BMP4 and BMP6 but not by BMP7. vivo ‘whole follicle’ context it is conceivable that the GC Under LH-stimulated conditions each BMP also increased compartment is relatively resistant to local BMP the abundance of noggin mRNA with the greatest increase signalling (BMP4, BMP6 and BMP7 from TC layer; being elicited by BMP6 (w20-fold: P!0.01). BMP2 mostly from GC layer) due to the presence of higher levels of extracellular BPs. As such, this preliminary evidence supports the concept of BMP signalling gradients across the follicle wall, akin to Discussion the morphogen gradients clearly implicated in key This study has provided novel information on the mRNA aspects of embryonic patterning such as dorsalisation, expression pattern of four different activin/BMP-BPs in segmentation and limb bud formation (Thomsen 1997, granulosal and thecal compartments of developing Balemans & Van Hul 2002, Chen et al. 2004, Kishigami bovine antral follicles. To our knowledge, such & Mishina 2005).

Reproduction (2011) 142 581–591 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access BMP-binding proteins and follicle development 585

more complex expression patterns but were all relatively low in 1–2 mm follicles. This raises the suggestion that gremlin has a specific role in these small, pre-selection antral follicles. Alternatively, local factors inductive for GC gremlin expression might be more abundant in small antral follicles. In this regard BMP2, BMP4 and GDF9 were found to increase gremlin expression by mouse GC in vitro (Pangas et al. 2004) and this study shows that BMP2, BMP4 and BMP6, but not BMP7, increase gremlin expression by cultured TC. However, as we report elsewhere (Glister et al. 2010) expression of BMP2, BMP4 and BMP7 showed little variation in bovine follicles from 1 to 8 mm in diameter. Moreover, expression of these BMPs by cultured TC and GC was unaffected by LH and FSH, respectively, even though IGF selectively reduced GC expression of BMP4. Whilst expression of all four BPs was considerably lower in TC than GC, when comparing LEA versus LEI follicles specifically, some differences are apparent in TC. Gremlin, chordin and follistatin were more highly expressed in the TC of LEI follicles compared with LEA, whilst noggin remained unchanged. This might imply a role in neutralising factors involved in maintenance of the dominant follicle, a proposal that should be amenable to in vitro testing using an RNAi ‘knockdown’ approach. To extend the above findings of spatiotemporal changes in follicular BMP-BP expression during antral follicle development, we conducted in vitro studies on isolated GC and TC to ascertain whether key trophic factors influenced expression of any of the four BPs, along with several key GC or TC markers. As anticipated, Figure 3 Immunohistochemical staining of bovine ovary sections and consistent with other studies (Campbell et al. 1996, showing the distribution of chordin, gremlin, noggin and follistatin Gutierrez et al. 1997, Glister et al. 2001, Marsters et al. immunoreactivity (brown) in the granulosa and theca layer of small 2003), treatment of GC with FSH enhanced E secretion antral follicles (w3 to 5 mm diameter). In each panel the left-hand image 2 shows an antibody-treated section while the right-hand image shows and promoted expression of CYP19, its own receptor an adjacent section treated with the appropriate non-immune serum (FSHR) and INHBA (encoding activin A subunit), or purified IgG. The scale bar in each right panel represents 50 mm. while IGF raised E2 secretion and CYP19 without effecting FSHR or INHBA expression. Follistatin mRNA The finding of differential expression patterns of expression was also raised by both treatments, consistent the four extracellular BPs throughout antral follicle with previous follistatin secretion data for FSH and IGF- progression suggests selectively regulated expression, treated bovine GC (Glister et al. 2001, 2003, 2006)and as well as different biological roles. For instance, GC with findings in other species (see Findlay 1993). expression of follistatin was much greater in LEA than LEI However, FSH had no effect on expression of the four follicles with the inverse seen for noggin. Notably, the BMPs examined or their antagonists – chordin, noggin previously reported GC expression patterns of INHBA and gremlin. Previously, in vivo PMSG treatment in mice (encoding activin A homodimer) and BMP2 in LEA was found to increase GC gremlin expression (Pangas versus LEI follicles (Glister et al. 2010) mirrored these et al. 2004). In contrast to the lack of effect of FSH on GC, changes. This supports the notion that activin induces treatment with IGF selectively reduced GC expression follistatin expression (Tano et al. 1995, Zhang et al. of BMP4, noggin and gremlin perhaps indicating a 1997, Silva & Knight 1998) while BMP2 induces noggin functional link between BMP4 and these two BPs. expression (Gazzerro & Minetti 2007), both acting in As anticipated, treatment of cultured TC with their an autoregulatory feedback manner to limit signalling key trophic factor, LH, promoted a significant (sixfold) by the respective ligands. increase in androgen secretion. However, LH had no Intriguingly, gremlin mRNA level was maximal in GC discernable effect on expression levels of any of the four of the smallest antral follicle stage examined (1–2 mm) BMPs or BPs analysed. Therefore, we infer that and then declined progressively as follicle size expression of BMPs and BMP-BPs by GC and TC from increased. In contrast, the other three BPs showed antral follicles is largely unaffected by gonadotrophin www.reproduction-online.org Reproduction (2011) 142 581–591

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access 586 C Glister and others

NS 2.5 BMP2 2.0 Chordin Control 2.0 1.5 NS 1.5 FSH 1.0 NS 1.0 NS 0.5 IGF 0.5 0 0 50 2.5 2.0 CYP19 ** BMP4 Noggin 40 2.0 1.5 NS 30 1.5 NS 1.0 20 * 1.0 0.5 * 10 0.5 ** 0 0 0 10 2.5 FSHR BMP6 2.0 Gremlin 8 2.0 ** 1.5 6 1.5 NS NS 1.0 NS 4 1.0

Relative mRNA abundance Relative * 2 NS 0.5 0.5 Figure 4 Effect of FSH and IGF1 analogue (IGF) on 0 0 0 the relative abundance of mRNA transcripts for 20 INHBA 2.5 BMP7 10 Follistatin NS ** CYP19, FSHR, INHBA, BMP2, BMP4, BMP6, 15 ** 2.0 8 BMP7, chordin, gremlin, noggin and follistatin in 1.5 6 10 NS primary cultures of bovine granulosa cells. Values NS 1.0 4 * are means and bars indicate S.E.M.(nZ4 inde- 5 0.5 2 pendent cultures). NS, not significant; *P!0.05 0 0 0 and **P!0.01 versus vehicle-treated controls. stimulation, at least under the defined culture conditions would be a robust model to investigate whether the used for this study. Further studies would be useful to different BMPs also modulated expression of any of the ascertain whether gonadotrophins and other factors four extracellular BPs. We first confirmed that BMP2, modulate TC/GC expression of BMP-BPs at other follicle BMP4, BMP6 and BMP7 drastically reduced expression stages, including preantral and luteinisation stages in of the key enzyme involved in androgen biosynthesis cattle, and indeed other species. CYP17 (O100-fold), with lesser effects on other Given the high responsiveness of cultured bovine TC components of the thecal steroidogenic pathway, to exogenous BMPs in terms of suppression of androgen including LHCGR (approximately sixfold), STAR secretion (Glister et al. 2005), we reasoned that this (approximately twofold), CYP11A (approximately

1.6 LHCGR 1.4 BMP P<0.0001 1.2 LH P=0.55 Basal 1.0 BMP × LH P=0.83 0.8 LH 0.6 0.4 0.2 0 5.0 4.5 STAR 1.8 HSD3B 4.0 BMP P<0.0001 1.6 BMP P<0.001 3.5 LH P=0.58 1.4 LH P=0.51 3.0 BMP × LH P=0.88 1.2 BMP × LH P=0.97 2.5 1.0 2.0 0.8 1.5 1.0 0.6 0.5 0.4 0 0.2 0 Relative mRNA abundance Relative 2.0 1.8 CYP11A1 CYP17 10 1.6 BMP P<0.0001 BMP P<0.0001 Figure 5 Effects of BMP2, BMP4, BMP6 and BMP7, alone 1.4 LH P=0.61 1 LH P=0.59 and in combination with LH, on the relative abundance 1.2 BMP × LH P=0.38 BMP × LH P=0.55 1.0 of five key steroidogenic mRNA transcripts (LHCGR, 0.8 0.1 0.6 STAR, CYP11A1, HSD3B and CYP17) in primary cultures 0.4 0.01 of bovine theca interna cells. Values are means and bars 0.2 Z 0 0.001 indicate S.E.M.(n 4 independent cultures). Results of VEH BMP2 BMP4 BMP6 BMP7 VEH BMP2 BMP4 BMP6 BMP7 two-way ANOVA are summarised.

Reproduction (2011) 142 581–591 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access BMP-binding proteins and follicle development 587

Chordin Gremlin Basal BMP P<0.001 BMP P<0.001 LH LH P=0.63 LH P=0.85 18 BMP × LH P=0.75 16 BMP × LH P=0.69 16 14 14 12 12 10 10 8 8 6 6 4 4 2 2 0 0 Noggin Follistatin BMP P<0.001 BMP P<0.73 LH P=0.39 LH P=0.84 30 3.0 BMP × LH P=0.06 BMP × LH P=0.98 Figure 6 Effects of BMP2, BMP4, BMP6 and BMP7, alone 25 2.5 and in combination with LH, on the relative abundance 20 2.0 of mRNA transcripts for chordin, gremlin, noggin and Relative mRNA abundance Relative 15 1.5 follistatin in primary cultures of bovine theca interna 10 1.0 cells. Values are means and bars indicate S.E.M. 5 0.5 (nZ4 independent cultures). Results of two-way ANOVA 0 0 VEH BMP2 BMP4 BMP6 BMP7 VEH BMP2 BMP4 BMP6 BMP7 are summarised. sixfold) and HSD3B (approximately twofold). Whilst all 2004) it would be anticipated that activin is a more four BMPs caused a modest increase in expression of effective inducer of follistatin expression and, indeed, chordin and gremlin, (independent of LH treatment) only activin-induced follistatin expression has been docu- one, BMP6, promoted a marked (greater than tenfold) mented for various cell-types including rat GC (Tano increase in noggin expression. None of the four BMPs et al. 1995), hepatocytes (Zhang et al. 1997) and bovine tested altered follistatin expression by TC, a finding at cumulus–oocyte complexes (Silva & Knight 1998), but variance with previous observations in bovine GC that not human granulosa–lutein cells (Tuuri & Ritovs 1995). show a clear BMP-induced upregulation of follistatin Apparent differences in the ability of individual BMPs secretion (Glister et al. 2004). to modulate expression of different BPs, evident from the Previous studies on various cell-types have reported studies referred to above, can partially be attributed to the ability of BMPs to induce expression of their own the different cell-types and species under investigation, antagonists. For instance, BMP2, BMP4 and BMP6 have and to differences in experimental design and culture each been shown to induce noggin expression by conditions. Taken together, however, it is increasingly osteoblasts, at the mRNA and protein level (Gazzerro evident that local autoregulatory loops operate at the & Minetti 2007) even though noggin appears to bind to target tissue/cell level with BMPs capable of inducing BMP2 and BMP4 with much greater affinity than to expression of their own extracellular antagonists. In turn, BMP6 (Pierre et al. 2005). Interestingly, the present data these act primarily as ‘ligand traps’ to sequester BMPs indicate that, of the four ligands we tested, only BMP6 and limit their signalling range by blocking interaction has the ability to induce noggin expression by bovine with signalling receptors (Reddi 2001, Gazzerro & TC. In a similar manner BMP2 and BMP4, which are Canalis 2006, Walsh et al. 2010). antagonised by gremlin, have been shown to induce Whilst it is well established that follistatin has an gremlin expression by mouse GC (Pangas et al. 2004) and rat osteoblasts (Pereira et al. 2000). In agreement, we intraovarian role in antagonising the effects of activin on report here that BMP2 and BMP4 can also increase TC, GC, luteal cells and oocytes (reviews: Findlay 1993, gremlin expression by bovine TC in vitro. BMP6 was Welt et al. 2002, Knight & Glister 2006) what evidence similarly effective at promoting gremlin expression is there to support a functional role of BMP-BPs in the whereas BMP7 had little or no effect. Chordin expression modulation of follicular function? Noggin has been has previously been reported to be non-inducible by shown to block the suppressive effect of BMP2 and BMPs (Canalis et al. 2003), despite binding directly to BMP4, but not BMP6, on progesterone secretion by and inhibiting BMP2, BMP4 and BMP7. However, in our sheep GC in vitro (Pierre et al. 2004). Likewise, gremlin study chordin expression was raised by all four BMPs and gremlin 2 (aka PRDC; protein related to DAN and tested. Previously, BMPs were shown to enhance cerberus) reversed the BMP2/BMP4-induced suppres- secretion of follistatin by both non-luteinised (Glister et sion of progesterone secretion by rat GC (Sudo et al. al. 2004) and luteinised (Kayani et al. 2009) bovine GC. 2004). Gremlin and chordin also antagonised the In contrast, this study found no effect of BMPs on TC suppressive action of BMP4 and BMP7, respectively, expression of follistatin mRNA. Since the binding affinity on androgen secretion by bovine TC in vitro (Glister of follistatin for activin is much greater than that for BMPs et al.2005). Moreover, in the absence of BMP (Otsuka et al. 2001, Amthor et al. 2002, Glister et al. treatments, both BPs promoted TC androgen secretion www.reproduction-online.org Reproduction (2011) 142 581–591

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access 588 C Glister and others indicating that endogenous BMPs negatively regulate collected, nZ5 pools analysed) and 5–6 mm (six follicles per androgen production in an autocrine/paracrine manner. pool; nZ6 pools collected, nZ5 pools analysed) size Given the marked inhibitory effect of activins and categories were combined for further analysis while all follicles BMPs on thecal androgen production, it has also been O7 mm in diameter were processed and analysed individually suggested (Glister et al.2005) that compromised (nZ7–9 per category). The number of follicle samples included intraovarian activin and/or BMP signalling, perhaps in the analysis was lower than the number collected because due to overexpression or activity of extracellular BPs only 80% of samples passed our quality control criteria based could by a contributory factor in polycystic ovarian on comparison of GC/TC-specific ‘marker’ transcript levels syndrome (PCOS), a condition associated with ovarian (see Glister et al. (2010) for further details). Isolated GC and TC androgen excess and impaired preovulatory follicle were homogenised in 0.5 ml of Tri-reagent (Sigma UK Ltd) and K development (Mason 2000). Consistent with this notion, stored at 80 8C for subsequent RNA purification. The size there have been several reports of raised circulating range of antral follicles investigated spanned key points in follistatin levels in women with PCOS (Norman et al. bovine follicular development, including ‘cyclic recruitment’ w 2001, Chen et al. 2009). However, we are not aware of at 3–5 mm, selection/deviation at 6to8mm,LHCGR any evidence implicating altered expression of other acquisition by GC at 9–10 mm and dominance at 11–18 mm extracellular BMP-BPs in the aetiology of PCOS. (Ireland et al. 2000, Mihm & Austin 2002, Ginther et al. 2003). In summary, this study has revealed that four different The final follicle size category was further subdivided based on follicular fluid E:P ratio reflecting presumptive healthy, secreted BMP/activin-BPs (gremlin, noggin, chordin and LEA dominant follicles (E:P ratio O1) and LEI, most likely follistatin) show divergent expression patterns in GC and undergoing regression (E:P ratio !1). TC compartments during bovine antral follicle develop- ment. With the exception of FSH-induced follistatin expression by GC, gonadotrophins do not appear to GC and TC culture modulate GC and TC expression of either BMPs or BMP- w BPs. In contrast, IGF induces follistatin expression whilst For in vitro experiments, GC and TC layers pooled from 50 follicles (4–6 mm diameter) per culture were collected as above suppressing BMP4, noggin and gremlin expression by and further processed as described by Glister et al. (2005) to GC. In addition, we have shown that treatment of TC obtain individual cell suspensions. The serum-free culture with BMPs can differentially regulate expression of their medium used was McCoy‘s 5A supplemented with 1% (v/v) own BPs, consistent with the notion of local autoregu- antibiotic–antimycotic solution, 10 ng/ml insulin (bovine latory feedback loop(s) controlling BMP signalling. pancreas), 2 mM L-glutamine, 10 mM HEPES, 5 mg/ml apo- Taken together, these findings support the hypothesis transferrin, 5 ng/ml sodium selenite and 0.1% (w/v) BSA. For that extracellular BPs play important roles in regulating GC cultures medium was supplemented with androstenedione K BMP signalling in ovarian follicles. Further studies are (10 7 M). Cells were plated at 75 000 viable cells/well in 96-well needed to delineate the relative contributions of plates (Nunclon, Life Technologies Ltd) and cultured for 6 days individual intraovarian BMP ligands and BPs in this at 38.5 8C. Media were removed every 48 h and replaced complex and dynamic organ. with fresh media containing treatments (see below). Conditioned media were stored at K20 8C for steroid immunoassays. At the end of culture viable cell number was determined using Materials and Methods neutral red assay (Campbell et al.1996, Glister et al.2001). Unless stated otherwise, all media and reagents were purchased from Sigma UK Ltd or Fisher Scientific Ltd RNA isolation from cultured cells (Loughborough, UK). In culture experiments in which total RNA was to be extracted for qRT-PCR analysis, cells were seeded into 24-well plates Ovaries and isolation of GC and TC for gene expression (0.5!106 cells/ml) with three replicate wells per treatment. At analysis the end of culture lysates were prepared using Tri-reagent and pooled lysates from replicate wells stored at K80 8C. As described in our recent companion paper (Glister et al. 2010) ovaries from non-pregnant cattle slaughtered at random stages of the oestrous cycle were collected from an abattoir and Culture treatments oestrous cycle stage (I–IV) was estimated by morphological appearance of the corpus luteum (Ireland et al. 1980). Only Ovine FSH (oFSH-19SIAPP) and LH (oLH-S-16) were provided ovaries judged to be from cattle in the mid- to late-luteal phase by the NHPP (Torrance, CA, USA), LR3-IGF-1 was purchased (stages II and III; days 5–17) were selected for follicle from Sigma and recombinant human (rh) BMP2, BMP4, BMP6 dissection. Follicles ranging in diameter from 1 to 18 mm and BMP7 were purchased from R&D Systems (Abingdon, UK). were dissected out, sorted by size and their GC, TC layers and Treatments were sterilised using 0.2 mm filters before further follicular fluid recovered for analysis. Individual follicles in the dilution in sterile culture medium. Treatments were added 1–2 mm (ten follicles per pool, nZ6 pools collected, nZ4 at 25 ml/well and an equal volume of blank medium added pools analysed), 3–4 mm (six follicles per pool; nZ7 pools to control wells.

Reproduction (2011) 142 581–591 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access BMP-binding proteins and follicle development 589

Steroid assays normalised to b-actin Ct value (uniform across experimental all groups: ANOVA PO0.2). For cell culture experiments the Concentrations of P were determined by ELISA (Sauer et al. 4 resultant DC values for each treatment replicate were then 1986, Bleach et al.2001). The detection limit was 20 pg/ml t normalised to the mean DC value of the respective vehicle- and intra- and inter-assay coefficient of variation (CV) were 8 and t treated control group. For tissue samples (GC and TC) DC 10% respectively. Concentrations of E were determined by RIA t 2 values for each transcript in a given sample were normalised to (Tannetta et al.1997) with a detection limit of 2 pg/ml and intra- the mean DC value for that transcript in all tissue samples. For and inter-assay CVof 6 and 9% respectively.Concentrations of A4 t graphical presentation DDCt values were converted to fold- were determined by RIA (Tannetta et al.1997) with a detection ðK Þ differences using the formula: fold-difference Z2 DDCt . limit of 100 pg/ml and intra- and inter-assay CV of 7 and 10% respectively. Immunohistochemistry RNA isolation, cDNA synthesis and real-time PCR Immunohistochemical staining was performed on 6 mm Total RNA was isolated from tissue samples and cultured cells sections of formalin-fixed, paraffin-embedded ovaries as using Tri-reagent as described previously (Glister et al. 2010). described previously (Glister et al. 2010). Ovaries from four Potential genomic DNA contamination was removed with an cattle were processed and each antibody was used to stain at RNase-free DNase kit (RQ1; Promega UK Ltd). RNA quantity least two sections from each of four different ovaries. After and quality were evaluated by spectrophotometry at 260/ microwave antigen retrieval using citrate buffer (10 mM, pH 6), 280 nm. cDNA was synthesised from 1 mg of RNA using the sections were blocked for 1 h using either 2.5% (v/v) horse Reverse-iT RT kit (Abgene, Epsom, UK; used according to serum (supplied with anti-mouse/rabbit ImmPRESS Kit, Vector manufacturer’s protocol) in a 20 ml reaction primed with Laboratories, Peterborough, UK) or 10% (v/v) bovine serum random hexamers. (used with biotinylated anti-goat IgG and Vectastain ABC Elite Primers (see Table 1) were designed using ABI PRISM primer kit; Vector Laboratories). The following primary antibodies (and express software (version 1.5; Perkin Elmer-Applied Biosys- appropriately matched control serum/purified IgG for adjacent tems, Warrington, UK) and validated using melt-curve analysis tissue section) were diluted in blocking buffer and incubated and agarose gel electrophoresis to verify that each primer pair with sections overnight at 4 8C: goat anti-mouse gremlin (R&D generated a single amplicon of the predicted size. cDNA Systems) at 5 mg/ml, goat anti-mouse chordin (R&D Systems) at template log-dilution curves were used to demonstrate 5 mg/ml, rabbit anti-human noggin (Abcam Ltd, Cambridge, satisfactory PCR efficiency and linearity. PCR assays were UK)at5mg/ml and rabbit anti-human follistatin (FP22; carried out in a volume of 24 ml, comprising 10 ml cDNA supplied by NIDDK, Baltimore, MD, USA) at 1:500 dilution. template (equivalent to w20 ng reverse-transcribed RNA), 1 ml For immunodetection of rabbit primary antibodies, the each forward and reverse primers (final concentration 0.4 mM) ImmPRESS Universal anti-mouse/rabbit IgG HRP polymer- and 12 ml QuantiTect SYBR Green QPCR 2! Master Mix based detection system was used with diaminobenzidine (Qiagen). Samples were processed on an ABI PRISM 7700 substrate according to the manufacturer’s instructions. For thermal cycler (Perkin Elmer-Applied Biosystems) with cycling immunodetection of goat primary antibodies, sections were conditions: 15 min at 95 8C (one cycle only) followed by 15 s incubated for 30 min with biotinylated bovine anti-goat IgG at 95 8C and 1 min at 60 8C for 40 cycles. (5 mg/ml; Santa Cruz, Insight Biotechnology, Wembley, UK) The DDCt method (Livak & Schmittgen 2001) was used before labelling with avidin–biotin HRP complex and diami- to compare the relative abundance of each mRNA transcript. nobenzidine substrate according to the supplier’s protocol. Ct values for each transcript in a given sample were first Slides were counterstained with haematoxylin, dehydrated and

Table 1 List of primers used for quantitative RT-PCR.

Amplicon Target Accession number Forward primer 50–30 Reverse primer 50–30 size (bp) FSHR NM_174061.1 GCCAGCCTCACCTACCCCAGC AATTGGATGAAGGTCAGAGGTTTGCC 75 LHCGR NM_174381.1 ATTGCCTCAGTCGATGCCCAGACC AAAAAGCCAGCCGCGCTGC 92 STAR NM_174189 TTTTTTCCTGGGTCCTGACAGCGTC ACAACCTGATCCTTGGGTTCTGCACC 103 CYP11A1 NM_176644 CAGTGTCCCTCTGCTCAACGTCC TTATTGAAAATTGTGTCCCATGCGG 99 HSD3B NM_174343.2 GCCACCTAGTGACTCTTTCCAACAGCG TGGTTTTCTGCTTGGCTTCCTCCC 111 CYP17 NM_174304 GACAAAGGCACAGACGTTGTGGTCA TGATCTGCAAGACGAGACTGGCATG 301 CYP19 NM_174305.1 CGCCACTGAGTTGATTTTTGCTGAGA TAAGGCTTTGCGCATGACCAGGTC 301 BMP2 XM_866011.1 CCAAGAGGCATGTGCGGATTAGCA TCCTTTCCCATCGTGGCCAAAAGT 101 BMP4 NM_001045877.1 TTTATGAGGTTATGAAGCCCCCGGC AGTTTCCCACCGCGTCACATTGTG 104 BMP6 XM_600972.2 GGCCCCGTTAACTCGACTGTGACAAA TTGAGGACGCCGAACAAAACAGGA 108 BMP7 XM_612246.2 TGCAAGATAGCCACTTCCTCACCGA GGGATCTTGGAGAGATCAAACCGGA 130 Chordin XM_001788437.1 CCTACCCGAATCCGCTTCTCTGACTCC GACAACCGAGGCACTGCCCGC 113 Gremlin NM_001082450.1 GAAGCGAGACTGGTGCAAAACCCA TATGCAACGGCACTGCTTGACACG 271 Noggin XM_582573.4 CAAGAAGCAGCGCCTGAGCAAGA GAAACAGCTGCCCACCTTCACGTAG 142 Follistatin NM_175801.2 TGAGCAAGGAGGAGTGTTGCAGCA CATCTGGCCTTGAGGAGTGCACATTC 301 ACTB NM_173979.3 ATCACCATCGGCAATGAGCGGTTC CGGATGTCGACGTCACACTTCATGA 128 www.reproduction-online.org Reproduction (2011) 142 581–591

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access 590 C Glister and others mounted under glass cover slips. Sections were viewed on a Gazzerro E & Canalis E 2006 Bone morphogenetic proteins and their Zeiss Axioskop 2 microscope (20! objective) and images antagonists. Reviews in Endocrine & Metabolic Disorders 7 51–65. acquired using a Zeiss Axiocam with Axiovision software. (doi:10.1007/s11154-006-9000-6) Gazzerro E & Minetti C 2007 Potential drug targets within bone morphogenetic protein signaling pathways. Current Opinion in Pharmacology 7 325–333. (doi:10.1016/j.coph.2007.01.003) Statistical analysis Ginther OJ, Beg MA, Donadeu FX & Bergfelt DR 2003 Mechanism of follicle deviation in monovular farm species. Animal Reproduction To reduce heterogeneity of variance immunoassay data were Science 78 239–257. (doi:10.1016/S0378-4320(03)00093-9) log-transformed before statistical analysis and QPCR data were Glister C, Tannetta DS, Groome NP & Knight PG 2001 Interactions analysed as DCt values before conversion to fold-difference between follicle-stimulating hormone and growth factors in modulating values. Results were evaluated using one- and/or two-way secretion of steroids and inhibin-related peptides by nonluteinized ANOVA and, where indicated, post hoc comparisons made by bovine granulosa cells. Biology of Reproduction 65 1020–1028. (doi:10. 1095/biolreprod65.4.1020) Fisher’s protected least significant difference (PLSD) test. Post Glister C, Groome NP & Knight PG 2003 Oocyte-mediated suppression of hoc tests were only made when ANOVA yielded a significant follicle-stimulating hormone- and insulin-like growth factor-induced F-ratio. In vitro results presented are based on combined data secretion of steroids and inhibin-related proteins by bovine granulosa from four independent culture experiments. cells in vitro: possible role of transforming growth factor alpha. Biology of Reproduction 68 758–765. (doi:10.1095/biolreprod.102.008698) Glister C, Kemp CF & Knight PG 2004 Bone morphogenetic protein (BMP) ligands and receptors in bovine ovarian follicle cells: actions of BMP-4, Declaration of interest -6 and -7 on granulosa cells and differential modulation of Smad-1 phosphorylation by follistatin. Reproduction 127 239–254. (doi:10. The authors declare that there is no conflict of interest that 1530/rep.1.00090) could be perceived as prejudicing the impartiality of the Glister C, Richards SL & Knight PG 2005 Bone morphogenetic proteins (BMP) -4, -6, and -7 potently suppress basal and luteinizing hormone- research reported. induced androgen production by bovine theca interna cells in primary culture: could ovarian hyperandrogenic dysfunction be caused by a defect in thecal BMP signaling? Endocrinology 146 1883–1892. (doi:10. 1210/en.2004-1303) Funding Glister C, Groome NP & Knight PG 2006 Bovine follicle development is This work was supported by the Biotechnology and Biological associated with divergent changes in activin-A, inhibin-A and follistatin and the relative abundance of different follistatin isoforms in follicular Sciences Research Council (grants BBS/B/10439 and fluid. Journal of Endocrinology 188 215–225. (doi:10.1677/joe.1.06485) BB/G017174/1 to P G Knight). Glister C, Satchell L & Knight PG 2010 Changes in expression of bone morphogenetic proteins (BMPs), their receptors and inhibin co-receptor betaglycan during bovine antral follicle development: inhibin can antagonize the suppressive effect of BMPs on thecal androgen production. Reproduction 140 699–712. (doi:10.1530/REP-10-0216) References Gutierrez CG, Campbell BK & Webb R 1997 Development of a long-term Amthor H, Christ B, Rashid-Doubell F, Kemp CF, Lang E & Patel K 2002 bovine granulosa cell culture system: induction and maintenance of Follistatin regulates bone morphogenetic protein-7 (BMP-7) activity to estradiol production, response to follicle-stimulating hormone, and stimulate embryonic muscle growth. Developmental Biology 243 morphological characteristics. Biology of Reproduction 56 608–616. 115–127. (doi:10.1006/dbio.2001.0555) (doi:10.1095/biolreprod56.3.608) Avsian-Kretchmer O & Hsueh AJ 2004 Comparative genomic analysis of Hsu DR, Economides AN, Wang X, Eimon PM & Harland RM 1998 The the eight-membered ring cystine knot-containing bone morphogenetic Xenopus dorsalizing factor Gremlin identifies a novel family of secreted protein antagonists. Molecular Endocrinology 18 1–12. (doi:10.1210/ proteins that antagonize BMP activities. Molecular Cell 1 673–683. me.2003-0227) (doi:10.1016/S1097-2765(00)80067-2) Balemans W & Van Hul W 2002 Extracellular regulation of BMP signaling Ireland JJ, Murphee RL & Coulson PB 1980 Accuracy of predicting stages in vertebrates: a cocktail of modulators. Developmental Biology 250 of bovine estrous cycle by gross appearance of the corpus luteum. 231–250. (doi:10.1006/dbio.2002.0779) Journal of Dairy Science 63 155–160. (doi:10.3168/jds.S0022-0302 Bleach EC, Glencross RG, Feist SA, Groome NP & Knight PG 2001 Plasma (80)82901-8) inhibin A in heifers: relationship with follicle dynamics, gonadotrophins, Ireland JJ, Mihm M, Austin E, Diskin MG & Roche JF 2000 Historical and steroids during the estrous cycle and after treatment with bovine perspective of turnover of dominant follicles during the bovine estrous follicular fluid. Biology of Reproduction 64 743–752. (doi:10.1095/ cycle: key concepts, studies, advancements, and terms. Journal of Dairy biolreprod64.3.743) Science 83 1648–1658. (doi:10.3168/jds.S0022-0302(00)75033-8) Campbell BK, Scaramuzzi RJ & Webb R 1996 Induction and maintenance Jasuja R, Allen BL, Pappano WN, Rapraeger AC & Greenspan DS 2004 of oestradiol and immunoreactive inhibin production with FSH by ovine Cell-surface heparan sulfate proteoglycans potentiate chordin antagon- granulosa cells cultured in serum-free media. Journal of Reproduction ism of bone morphogenetic protein signaling and are necessary for and Fertility 106 7–16. (doi:10.1530/jrf.0.1060007) cellular uptake of chordin. Journal of Biological Chemistry 279 Canalis E, Economides AN & Gazzerro E 2003 Bone morphogenetic 51289–51297. (doi:10.1074/jbc.M408129200) proteins, their antagonists, and the skeleton. Endocrine Reviews 24 Juengel JL & McNatty KP 2005 The role of proteins of the transforming 218–235. (doi:10.1210/er.2002-0023) growth factor-beta superfamily in the intraovarian regulation of follicular Chen D, Zhao M & Mundy GR 2004 Bone morphogenetic proteins. Growth development. Human Reproduction Update 11 143–160. (doi:10.1093/ Factors 22 233–241. (doi:10.1080/08977190412331279890) humupd/dmh061) Chen MJ, Chen HF, Chen SU, Ho HN, Yang YS & Yang WS 2009 The Kameda T, Koike C, Saitoh K, Kuroiwa A & Iba H 1999 Developmental relationship between follistatin and chronic low-grade inflammation patterning in chondrocytic cultures by morphogenic gradients: BMP in women with polycystic ovary syndrome. Fertility and Sterility 92 induces expression of Indian hedgehog and noggin. Genes to Cells 4 2041–2044. (doi:10.1016/j.fertnstert.2009.06.009) 175–184. (doi:10.1046/j.1365-2443.1999.00250.x) Findlay JK 1993 An update on the roles of inhibin, activin, and follistatin as Kayani AR, Glister C & Knight PG 2009 Evidence for an inhibitory role of local regulators of folliculogenesis. Biology of Reproduction 48 15–23. bone morphogenetic protein(s) in the follicular–luteal transition in cattle. (doi:10.1095/biolreprod48.1.15) Reproduction 137 67–78. (doi:10.1530/REP-08-0198)

Reproduction (2011) 142 581–591 www.reproduction-online.org

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access BMP-binding proteins and follicle development 591

Khokha MK, Hsu D, Brunet LJ, Dionne MS & Harland RM 2003 Gremlin is Sasai Y, Lu B, Steinbeisser H & De Robertis EM 1995 Regulation of neural the BMPantagonist required for maintenance of Shh and Fgf signals during induction by the Chd and Bmp-4 antagonistic patterning signals in limb patterning. Nature Genetics 34 303–307. (doi:10.1038/ng1178) Xenopus. Nature 377 757. (doi:10.1038/377757a0) Kishigami S & Mishina Y 2005 BMP signaling and early embryonic Sauer MJ, Foulkes JA, Worsfold A & Morris BA 1986 Use of progesterone patterning. Cytokine & Growth Factor Reviews 16 265–278. (doi:10. 11-glucuronide-alkaline phosphatase conjugate in a sensitive microtitre- 1016/j.cytogfr.2005.04.002) plate enzymeimmunoassay of progesterone in milk and its application to Knight PG & Glister C 2006 TGF-beta superfamily members and ovarian follicle pregnancy testing in dairy cattle. Journal of Reproduction and Fertility 76 development. Reproduction 132 191–206. (doi:10.1530/rep.1.01074) 375–391. (doi:10.1530/jrf.0.0760375) Lamb TM, Knecht AK, Smith WC, Stachel SE, Economides AN, Stahl N, Shimasaki S, Moore RK, Otsuka F & Erickson GF 2004 The bone Yancopolous GD & Harland RM 1993 Neural induction by the secreted morphogenetic protein system in mammalian reproduction. Endocrine polypeptide noggin. Science 262 713–718. (doi:10.1126/science.8235591) Reviews 25 72–101. (doi:10.1210/er.2003-0007) Livak KJ & Schmittgen TD 2001 Analysis of relative gene expression data Shimonaka M, Inouye S, Shimasaki S & Ling N 1991 Follistatin binds to using real-time quantitative PCR and the 2(KDelta Delta C(T)) method. both activin and inhibin through the common subunit. Endocrinology Methods 25 402–408. (doi:10.1006/meth.2001.1262) 128 3313–3315. (doi:10.1210/endo-128-6-3313) Marsters P, Kendall NR & Campbell BK 2003 Temporal relationships Silva CC & Knight PG 1998 Modulatory actions of activin-A and follistatin between FSH receptor, type 1 insulin-like growth factor receptor, on the developmental competence of in vitro-matured bovine oocytes. and aromatase expression during FSH-induced differentiation of bovine Biology of Reproduction 58 558–565. (doi:10.1095/biolreprod58.2.558) granulosa cells maintained in serum-free culture. Molecular and Cellular Smith WC, Knecht AK, Wu M & Harland RM 1993 Secreted noggin protein Endocrinology 203 117–127. (doi:10.1016/S0303-7207(03)00093-5) mimics the Spemann organizer in dorsalizing Xenopus mesoderm. Mason H 2000 Function of the polycystic ovary. Human Fertility 3 80–85. Nature 361 547–549. (doi:10.1038/361547a0) (doi:10.1080/1464727002000198741) Sudo S, Avsian-Kretchmer O, Wang LS & Hsueh AJ 2004 Protein related Michos O, Panman L, Vintersten K, Beier K, Zeller R & Zuniga A 2004 to DAN and cerberus is a bone morphogenetic protein antagonist that Gremlin-mediated BMP antagonism induces the epithelial–mesenchy- participates in ovarian paracrine regulation. Journal of Biological mal feedback signaling controlling metanephric kidney and limb Chemistry 279 23134–23141. (doi:10.1074/jbc.M402376200) organogenesis. Development 131 3401–3410. (doi:10.1242/dev.01251) Sugino K, Kurosawa N, Nakamura T, Takio K, Shimasaki S, Ling N, Titani K Mihm M & Austin EJ 2002 The final stages of dominant follicle selection & Sugino H 1993 Molecular heterogeneity of follistatin, an activin- in cattle. Domestic Animal Endocrinology 23 155–166. (doi:10.1016/ binding protein. Higher affinity of the carboxyl-terminal truncated forms S0739-7240(02)00153-4) for heparan sulfate proteoglycans on the ovarian granulosa cell. Miyazono K, Kusanagi K & Inoue H 2001 Divergence and convergence of Journal of Biological Chemistry 268 15579–15587. TGF-beta/BMP signaling. Journal of Cellular Physiology 187 265–276. Sugino H, Sugino K, Hashimoto O, Shoji H & Nakamura T 1997 Follistatin (doi:10.1002/jcp.1080) and its role as an activin-binding protein. Journal of Medical Nakamura T, Takio K, Eto Y, Shibai H, Titani K & Sugino H 1990 Activin- Investigation 44 1–14. binding protein from rat ovary is follistatin. Science 247 836–838. Tannetta DS, Fray MD, Wrathall JH, Bleach EC, Glencross RG & Knight PG (doi:10.1126/science.2106159) 1997 Effects of supplementary treatment with bovine growth hormone on Norman RJ, Milner CR, Groome NP & Robertson DM 2001 Circulating hormonal and ovulatory responses to inhibin immunization in ewes. follistatin concentrations are higher and activin concentrations are lower Journal of Reproduction and Fertility 110 255–262. (doi:10.1530/jrf.0. in polycystic ovarian syndrome. Human Reproduction 16 668–672. 1100255) (doi:10.1093/humrep/16.4.668) Tano M, Minegishi T, Nakamura K, Nakamura M, Karino S, Miyamoto K & Otsuka F, Moore RK, Iemura S, Ueno N & Shimasaki S 2001 Follistatin Ibuki Y 1995 Regulation of follistatin messenger ribonucleic acid in inhibits the function of the oocyte-derived factor BMP-15. Biochemical cultured rat granulosa cells. Molecular and Cellular Endocrinology 109 and Biophysical Research Communications 289 961–966. (doi:10.1006/ 167–174. (doi:10.1016/0303-7207(95)03499-W) bbrc.2001.6103) Pangas SA, Jorgez CJ & Matzuk MM 2004 Growth differentiation factor 9 Thomsen GH 1997 Antagonism within and around the organizer: BMP regulates expression of the bone morphogenetic protein antagonist inhibitors in vertebrate body patterning. Trends in Genetics 13 209–211. gremlin. Journal of Biological Chemistry 279 32281–32286. (doi:10. (doi:10.1016/S0168-9525(97)01117-7) 1074/jbc.M403212200) Tuuri T & Ritovs O 1995 Regulation of the activin-binding protein follistatin Pereira RC, Rydziel S & Canalis E 2000 Bone morphogenetic protein-4 cultured human luteinizing granulosa cells: characterization of the regulates its own expression in cultured osteoblasts. Journal of Cellular effects of follicle stimulating hormone, prostaglandin E2, and different Physiology 182 239–246. (doi:10.1002/(SICI)1097-4652(200002)182: growth factors. Biology of Reproduction 53 1508–1516. (doi:10.1095/ 2!239::AID-JCP13O3.0.CO;2-W) biolreprod53.6.1508) Piccolo S, Sasai Y, Lu B & De Robertis EM 1996 Dorsoventral patterning in Walsh DW, Godson C, Brazil DP & Martin F 2010 Extracellular BMP- Xenopus: inhibition of ventral signals by direct binding of chordin to antagonist regulation in development and disease: tied up in knots. BMP-4. Cell 86 589–598. (doi:10.1016/S0092-8674(00)80132-4) Trends in Cell Biology 20 244–256. (doi:10.1016/j.tcb.2010.01.008) Pierre A, Pisselet C, Dupont J, Mandon-Pepin B, Monniaux D, Monget P & Welt C, Sidis Y, Keutmann H & Schneyer A 2002 Activins, inhibins, and Fabre S 2004 Molecular basis of bone morphogenetic protein-4 inhibitory follistatins: from endocrinology to signaling. A paradigm for the new action on progesterone secretion by ovine granulosa cells. Journal of millennium. Experimental Biology and Medicine 227 724–752. Molecular Endocrinology 33 805–817. (doi:10.1677/jme.1.01545) Zhang YQ, Kanzaki M, Shibata H & Kojima I 1997 Regulation of the Pierre A, Pisselet C, Monget P, Monniaux D & Fabre S 2005 Testing expression of follistatin in rat hepatocytes. Biochimica et Biophysica Acta the antagonistic effect of follistatin on BMP family members in ovine 1354 204–210. (doi:10.1016/S0167-4781(97)00085-7) granulosa cells. Reproduction, Nutrition, Development 45 419–425. Zimmerman LB, De Jesus-Escobar JM & Harland RM 1996 The Spemann (doi:10.1051/rnd:2005031) organizer signal noggin binds and inactivates bone morphogenetic Reddi AH 2001 Bone morphogenetic proteins: from basic science to protein 4. Cell 86 599–606. (doi:10.1016/S0092-8674(00)80133-6) clinical applications. Journal of Bone and Joint Surgery. American Zuniga A, Haramis AP, McMahon AP & Zeller R 1999 Signal relay by BMP Volume 83-A (Supplement 1) S1–S6. antagonism controls the SHH/FGF4 feedback loop in vertebrate limb Robertson DM, Klein R, de Vos FL, McLachlan RI, Wettenhall RE, Hearn MT, buds. Nature 401 598–602. (doi:10.1038/44157) Burger HG & de Kretser DM 1987 The isolation of polypeptides with FSH suppressing activity from bovine follicular fluid which are structurally different to inhibin. Biochemical and Biophysical Research Communi- Received 4 May 2011 cations 149 744–749. (doi:10.1016/0006-291X(87)90430-X) First decision 2 June 2011 Sasai Y, Lu B, Steinbeisser H, Geissert D, Gont LK & De Robertis EM 1994 Xenopus chordin: a novel dorsalizing factor activated by organizer-specific Revised manuscript received 2 August 2011 homeobox genes. Cell 79 779–790. (doi:10.1016/0092-8674(94)90068-X) Accepted 5 August 2011 www.reproduction-online.org Reproduction (2011) 142 581–591

Downloaded from Bioscientifica.com at 09/29/2021 12:39:17AM via free access